DATES OF CONRAD. 
229 
entries. Sometimes when there were a few diagnoses over 
enough to fill a “signature” of 8 or 16 pages, the excess 
was printed on the cover, and sometimes the cover of one 
and the same part was twice surcharged ; in this way some 
covers have fewer diagnoses than are to be found on other 
presumably later, copies. Mr. Conrad’s extraordinary and 
habitual carelessness, or want of memory, which grew upon 
him, especially in later years, to such an extent that he 
finally decided to attempt no more work, was a marked 
factor in inducing variations. It would seem from the dif¬ 
ferences observed in the various copies I have examined 
that the sheets were kept on hand and made up when called 
for from time to time by subscribers. The edition of a plate 
did not always hold out, and in nearly every copy there is 
at least one plate which is not the original but a substitute. 
For some of the plates Mr. Conrad used the copper plates 
engraved by Lesueur for the first series of the Journal of 
the Academy of Natural Sciences, simply stopping out the 
Academy’s heading on his prints. Sometimes in copying a 
plate on the lithographic stone to supply a deficiency the 
new copy would differ in shading or figures would be added 
which were not on the first edition of it. There are fre¬ 
quent cases where the same plate is found in two states 
with two different printers’ names at the bottom, though the 
copies are otherwise exactly alike. In several cases the 
second edition of a plate is wrongly numbered, so that in 
the copy containing it there will be two plates with the same 
number and one apparently deficient. The plates, if cor¬ 
rectly numbered and all present, should number for the 
four parts 1-32 and. 34-49; there never was any plate 33. 
The subscribers in binding did not always save the covers. 
It will therefore not surprise the reader to learn that the 
utmost search has not revealed an absolutely perfect and 
complete copy of the work nor any two exactly similar 
copies of it. Nevertheless, the differences are not of very 
great importance. Part 1 was marked by Conrad as issued 
January, 1838. It came out with a plain cover, so far as 
